Visual Question Answering Instruction: Unlocking Multimodal Large Language Model To Domain-Specific Visual Multitasks
This work addresses the challenge of adapting MLLMs for domain-specific visual tasks, which require more explicit visual understanding, representing an incremental advancement in multimodal AI applications.
The authors tackled the problem of extending multimodal large language models (MLLMs) to domain-specific visual tasks by developing Visual Question Answering Instruction (VQA-IN), a method to convert datasets into a unified question-answering format. The result was that this approach achieved high scores on domain-specific tasks while maintaining performance on vision-language tasks in a multitask setting.
Having revolutionized natural language processing (NLP) applications, large language models (LLMs) are expanding into the realm of multimodal inputs. Owing to their ability to interpret images, multimodal LLMs (MLLMs) have been primarily used for vision-language tasks. Currently, MLLMs have not yet been extended for domain-specific visual tasks, which require a more explicit understanding of visual information. We developed a method to transform domain-specific visual and vision-language datasets into a unified question answering format called Visual Question Answering Instruction (VQA-IN), thereby extending MLLM to domain-specific tasks. The VQA-IN was applied to train multiple MLLM architectures using smaller versions of LLMs (sLLMs). The experimental results indicated that the proposed method achieved a high score metric on domainspecific visual tasks while also maintaining its performance on vision-language tasks in a multitask manner.